---
title: "Latest reads"
description: "The engineering internet, summarised so you can actually read it."
---

# Latest reads
> The engineering internet, summarised so you can actually read it.

## Articles

### [ThunderAgent: 2x Faster Agentic Inference for Synthetic Data Generation at Scale](https://yomu.fyi/post/thunderagent-2x-faster-agentic-inference-for-synthetic-data-generation.md)
- Company: [Together AI](https://yomu.fyi/company/together-ai.md)
- Published: Jul 29, 2026
- Type: Announcement

ThunderAgent is a program-level scheduler for high-throughput agentic inference, developed by researchers at Georgia Tech, UIUC, CMU, and Together AI. It solves KV cache thrashing, which occurs when request-level engines like vLLM and SGLang evict an agent’s KV cache during tool calls and recompute it on resume, creating a cascade of evictions. ThunderAgent tracks each workflow as a schedulable program, pausing low-priority workflows under memory pressure and routing resumed programs through a global waiting queue to the node with the most available capacity. In synthetic data generation pipelines, it achieves up to 2.5x single-node throughput and 2.4x speedup on 8 nodes with near-linear scaling, and was accepted as an ICML 2026 Spotlight paper.


### [Building an agentic harness that outlasts the model](https://yomu.fyi/post/building-an-agentic-harness-that-outlasts-the-model.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Jul 29, 2026
- Type: Explainer

Shopify’s Application Security team built an agentic harness that scans software for vulnerabilities, proves candidates with real tests, and opens Shopify-specific remediation pull requests. Its Dispatch orchestrator partitions repositories, runs category-specific Hunting agents in parallel, verifies findings sequentially with a different model, and persists architecture artifacts for later diff-based scans. Test oracles require realistic integration or functional coverage, including cross-tenant fixtures for IDOR checks, while unproven candidates are rejected or downgraded. Across more than 80 applications and thousands of scans, the workflow produced over 300 findings, including two rated Critical by Shopify’s severity calculator, with full scans costing $50–$300 and incremental scans $5–$50. The post concludes that durable value lies in the harness’s verification, partitioning, cross-model review, and deterministic infrastructure rather than in any single model.


### [How NorthStar Anesthesia built a scheduling app for a workforce of 3,000 clinicians in weeks](https://yomu.fyi/post/how-northstar-anesthesia-built-a-scheduling-app-for-a-workforce-of-3-0.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Evan Pandya
- Published: Jul 28, 2026
- Type: Problem & solution

NorthStar Anesthesia needed a better way for roughly 3,000 physicians and Certified Registered Nurse Anesthetists (CRNAs) to coordinate shifts, because its commercial scheduling platform hid colleagues’ time-off data and its dashboard pilot was not mobile-friendly. Synaptiq built a React TypeScript scheduling app with one software engineer and deployed it through Databricks Apps in a few weeks, using the existing Databricks data foundation, governance, and Microsoft Entra ID single sign-on. The app provides color-coded views by clinician type, facility selection, day/week/month calendars, shift notes, search and filtering, and time-off visibility, with data refreshing every 30 minutes. Daily unique users rose from 75–80 at launch to more than 110 as the first clinician group migrated, while users reported less stress; planned additions include natural-language queries, push notifications, and automated staffing reports.


### [Developing Healthcare Robotics with GPU-Native Medical Physics Simulation](https://yomu.fyi/post/developing-healthcare-robotics-with-gpu-native-medical-physics-simulat.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Michelle Horton
- Published: Jul 28, 2026
- Type: Announcement

Healthcare robotics faces data scarcity, long-tail clinical scenarios, and multi-year development cycles due to reliance on physical phantoms and animal studies. To resolve this, NVIDIA Medical Physics Simulation introduces GPU-native simulation within NVIDIA Isaac for Healthcare, integrating classical physics solvers and generative world models on a single GPU. The Endoluminal Simulation Module models flexible instruments as Cosserat rods solved via globally coupled extended position-based dynamics, maintaining zero-copy PyTorch interoperability via NVIDIA Warp. Concurrently, Cosmos-H-Dreams leverages distilled video diffusion models via the FlashDreams runtime to stream action-conditioned surgical video at over 30 frames per second. This dual-track architecture enables synchronized physics, synthetic fluoroscopic rendering, and scaled reinforcement learning policies across hundreds of parallel GPU environments.


### [The audience is the asset. Media finance teams need to understand them to protect the margin.](https://yomu.fyi/post/the-audience-is-the-asset-media-finance-teams-need-to-understand-them.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Elena Tesser
- Published: Jul 28, 2026
- Type: Explainer

Media finance teams are tasked with understanding how audience value flows across subscriptions, advertising, and content investment so the business can protect margin. Streaming has turned one wholesale audience into multiple monetization strategies, while ad-supported tiers now account for 59% of new streaming sign-ups, making timely measurement and pricing analysis more important. Ontology preserves the meaning and context of figures as audiences, titles, channels, and business conditions change, distinguishing a correct answer from one that is merely accurate. Databricks Genie is described as a governed, data-smart AI coworker that answers sourced natural-language questions, learns from interactions, and shows its work; Genie-powered apps let hundreds of DIRECTV analysts and leaders query more than 1,200 customer-level attributes, while people retain decision authority.


### [Chief data officer: role, responsibilities, and career guide](https://yomu.fyi/post/chief-data-officer-role-responsibilities-and-career-guide.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Databricks Staff
- Published: Jul 28, 2026
- Type: Explainer

The chief data officer (CDO) is a senior executive who treats enterprise data as a strategic asset, overseeing data strategy, management, quality, governance, and its use in analytics and AI. The role sits between business strategy, information technology, and data science, while distinguishing the CDO’s focus on data from the CIO’s responsibility for technology infrastructure. Responsibilities span the data lifecycle, governance frameworks, access controls, quality metrics, business intelligence, analytics prioritization, and organizational data capability. The role has shifted from compliance-focused stewardship toward enterprise AI strategy, with enablement-oriented governance making reliable, AI-ready data accessible while protecting its integrity. The guide also presents executive authority, data literacy, and leadership judgment as important conditions for delivering measurable business value.


### [Eval-driven development: Lessons from evaluating GenAI at scale](https://yomu.fyi/post/eval-driven-development-lessons-from-evaluating-genai-at-scale.md)
- Company: [Airbnb](https://yomu.fyi/company/airbnb.md)
- Author: Rohit Girme
- Published: Jul 28, 2026

Airbnb outlines its framework for eval-driven development (EDD), treating GenAI evaluation as a continuous engineering discipline rather than an afterthought. The strategy combines programmatic checks, calibrated LLM-as-a-judge evaluators, and human review to detect subtle quality regressions. By inspecting intermediate agentic execution traces and real-world failure modes, teams ensure generative models remain reliable and aligned with product requirements.


### [From prototype to production: High QPS for Databricks AI Search](https://yomu.fyi/post/from-prototype-to-production-high-qps-for-databricks-ai-search.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Adam Gurary, Sheng Zhan, Ankit Vij, Vadim Antonov, Yu-Ju Huang, Dima Kotlyarov
- Published: Jul 28, 2026
- Type: Announcement

Databricks AI Search now offers generally available high-QPS scaling for Standard endpoints, addressing production workloads such as search bars, recommendations, and real-time entity resolution. Operators set a human-readable target\_qps value when creating or updating an endpoint through the Python SDK, REST API, or UI, while Databricks provisions the required compute without manual replica counts, node sizing, or load balancers. Existing Unity Catalog governance and Delta Sync remain in place, and scaling progress is exposed through scaling\_info as it moves from SCALING\_CHANGE\_IN\_PROGRESS to SCALING\_CHANGE\_APPLIED; endpoint observability shows requests per second, latency, and health. Service principal authentication is recommended for high-QPS traffic, while personal access tokens are capped at a few tens of QPS, and new capacity applies when an index is created or synced.


### [How Databricks manages its own coding agent spend with Unity AI Gateway Budgets](https://yomu.fyi/post/how-databricks-manages-its-own-coding-agent-spend-with-unity-ai-gatewa.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Rohit Agrawal, Shuyu Cao, Darming Zhao, Zack Siegel, Aaron Davidson
- Published: Jul 28, 2026
- Type: Problem & solution

Databricks describes how it manages coding-agent costs as thousands of engineers use Claude Code, Codex, Cursor, and other tools, creating exposure to runaway automation and growing R&D spend. The company routes all agent traffic through Unity AI Gateway Budgets and separates short-term runaway-spend protection from long-term monthly spend governance. A daily budget triggers self-service acknowledgement through Slack, an internal portal, or the CLI, while a high monthly limit uses manager-approved, project-scoped tiers that expire. Both budgets apply simultaneously, so effective usage is capped by the lower of the month-to-date total plus one runaway increment and the monthly maximum. Centralizing metering also gives managers and finance shared usage data, and the company reports that approval queues disappeared, monthly requests became rare, and engineers stopped rationing usage.


### [Our position on open-weights models](https://yomu.fyi/post/our-position-on-open-weights-models.md)
- Company: [Anthropic](https://yomu.fyi/company/anthropic.md)
- Published: Jul 28, 2026
- Type: Explainer

Anthropic CEO Dario Amodei states that Anthropic has never advocated banning open-weights models, while distinguishing models without dangerous capabilities as a public good from risks associated with sufficiently capable systems. The post identifies two concerns: authoritarian governments gaining military superiority or enabling repression, and powerful models being misused for cyberattacks, biological attacks, or serious alignment problems. It argues open weights can increase misuse risk because guardrails and monitoring are difficult after release, but says banning their use by US businesses would not address bad actors. Instead, Amodei supports restricting powerful chips and chipmaking equipment exports to China, disrupting industrial-scale distillation, and requiring global pre-release safety testing for sufficiently capable open and closed models. He agrees open weights can expand access, competition, and customer control, but says claims about safeguards should be tested empirically.


### [Coding Agent Horror Stories: The 29 Million Secret Problem](https://yomu.fyi/post/coding-agent-horror-stories-the-29-million-secret-problem.md)
- Company: [Docker](https://yomu.fyi/company/docker.md)
- Author: Jennifer Kohl
- Published: Jul 28, 2026
- Type: Problem & solution

Compromised versions of the Nx npm package, dubbed the s1ngularity campaign, used a post-install hook to repurpose installed AI coding agents as credential scanners. Because AI command-line interfaces often run with the developer's full user permissions, the malware invoked tools such as Claude Code, Gemini CLI, and Amazon Q with permission-bypass flags to map sensitive files. GitGuardian findings indicate that AI-assisted code leaks secrets at roughly twice the baseline rate, largely because models ingest live credentials into working contexts. Docker Sandboxes addresses this risk by restricting the execution environment to the project workspace and keeping actual credentials on the host. When sandboxed with sbx, agents receive sentinel placeholders, while an egress proxy injects real secrets into outgoing network requests.


### [How Vercel used Cursor to build Queues](https://yomu.fyi/post/how-vercel-used-cursor-to-build-queues.md)
- Company: [Cursor](https://yomu.fyi/company/cursor.md)
- Author: Cursor
- Published: Jul 28, 2026
- Type: Explainer

Vercel built its durable event streaming system, Vercel Queues, entirely with Cursor, using agents across three rewrites and validating changes against real AWS environments. Engineers had agents work on the Queues API, connect to AWS, inspect DynamoDB and Kinesis side effects, inject faults, simulate bugs, and verify results, while Cursor retained context through long sessions and compaction events. The workflow also supported model switching, parallel planning agents, inline review, stacked PRs, and asynchronous Cloud Agents. Among Vercel’s Cursor power users, PR throughput increased 54%, PR cycle time decreased 89%, resources per feature fell 27%, and the company reports 104 hours saved per developer annually. Product managers also use Cursor for repository analysis, product analytics, and small fixes without drawing on engineering resources.


### [Introducing Cursor Start](https://yomu.fyi/post/introducing-cursor-start.md)
- Company: [Cursor](https://yomu.fyi/company/cursor.md)
- Author: Cursor
- Published: Jul 28, 2026
- Type: Announcement

Cursor has launched Cursor Start, a monthly plan for developers in India priced at ₹649, with tax included and payment through UPI or card. The plan responds to requests for India-specific pricing and UPI support, and follows a reported tripling of Cursor’s Indian user base to more than 3M developers, making India its third-largest market globally. Start provides generous access to Grok 4.5 and Composer, more agent requests than the Free plan across desktop, web, iOS, and CLI, plus always-on cloud agents, Cursor for iOS, and workflow extensions through plugins, MCP servers, hooks, and skills. It sits between Free and Pro, offering everyday building capabilities while Pro adds broader model access and features such as Bugbot, Auto mode, Automations, and the Cursor SDK.


### [Build a Flask App with Okta for Secure OIDC Login and Authorized API Calls](https://yomu.fyi/post/build-a-flask-app-with-okta-for-secure-oidc-login-and-authorized-api-c.md)
- Company: [Okta](https://yomu.fyi/company/okta.md)
- Author: Akanksha Bhasin
- Published: Jul 28, 2026
- Type: Tutorial

Flask web applications require dedicated mechanisms to authenticate users securely and authorize access to downstream services. Developers can address this by integrating Okta as an OpenID Connect identity provider using the Authlib library. Authlib registers Okta as a remote provider and automates the Authorization Code flow using Proof Key for Code Exchange with SHA-256 code challenge verification. After logging in through Okta and completing the callback route, the Flask web application uses the acquired OAuth 2.0 access token to query a protected backend resource server. This architecture ensures that user credentials remain managed by Okta while client requests to backend API endpoints carry valid, scoped authorization tokens.


### [Get Started with Genie One: Top AI Cowork Use Cases for Business Users](https://yomu.fyi/post/get-started-with-genie-one-top-ai-cowork-use-cases-for-business-users.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Cynthya Peranandam, Sydney Sundell
- Published: Jul 28, 2026
- Type: Tutorial

Genie One is presented as a data-smart, agentic coworker for business users who need work completed across calendars, CRMs, data warehouses, ticketing systems, and shared documents rather than through a standalone chatbot. The guide explains four workflows: recurring business reviews, meeting preparation and follow-up, knowledge-work document automation, and operational monitoring with alerts. Users connect relevant sources, provide templates or natural-language rules, create reusable skills, test outputs, schedule routines, and refine instructions over time; the monitoring workflow can compare metrics with defined thresholds and email alerts. The post recommends starting with one narrow recurring task, connecting its required sources securely with data or IT partners, and iterating toward a library of saved skills that can be used across the organization.


### [The EU Digital Product Passport: a traceability deadline](https://yomu.fyi/post/the-eu-digital-product-passport-a-traceability-deadline.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Daniel Dahlin
- Published: Jul 27, 2026
- Type: Explainer

The EU’s Ecodesign for Sustainable Products Regulation (ESPR) will require in-scope products sold in the single market to carry a machine-readable Digital Product Passport (DPP), with batteries first from February 2027. The passport links a unique identifier and data carrier to lifecycle information that must remain accurate, while operators retain responsibility for data stored on their own backend, tiered access, supplier inputs, and any Registry registration. Implementation is primarily a data-integration and governance problem, spanning tier-N suppliers, per-unit operational records, lineage, analytics, AI, and controlled sharing rather than QR-code generation. Databricks components including Lakebase, Unity Catalog, Lakeflow Spark Declarative Pipelines, Databricks Apps, and OpenSharing are presented with a battery-focused Solution Accelerator as an operator-side reference architecture. The stated benefits extend beyond market access to faster recalls, sourcing-risk visibility, and lineage-based sustainability reporting, but technology alone does not establish compliance.


### [Building Reliable Production AI with Durable Workflows](https://yomu.fyi/post/building-reliable-production-ai-with-durable-workflows.md)
- Company: [Salesforce](https://yomu.fyi/company/salesforce.md)
- Author: Scott Nyberg
- Published: Jul 27, 2026
- Type: Problem & solution

AI prototypes rely on simple prompt-response interactions, but production AI systems handling thousands of records must manage distributed execution issues like worker crashes, rate limits, and transient failures. Rather than treating long-running operations as single large jobs or rebuilding state from logs, systems require durable workflows that preserve execution history across failures. In Agentforce Grid, execution is organized into a hierarchy where column runs divide into recoverable batches and rows composed of discrete activities. Each batch checkpoints after every successful row, retrying up to ten times with exponential backoff and resuming strictly at incomplete work to prevent duplicated model calls. Exposing progress across these hierarchical boundaries ensures that both users and engineers can distinguish completed, failing, and retryable units.


### [Indexing the Data Lake for Online Point Queries](https://yomu.fyi/post/indexing-the-data-lake-for-online-point-queries.md)
- Company: [Spotify](https://yomu.fyi/company/spotify.md)
- Author: Spotify Engineering
- Published: Jul 27, 2026
- Type: Problem & solution

Online services and AI agents require low-latency point queries over massive datasets that are economically impractical to store entirely in key-value systems like Bigtable or DynamoDB. While cloud object storage offers rapid request latency, traditional distributed query engines like Trino and BigQuery introduce seconds of planning and scheduling overhead along with dependent file reads. Random Access Parquet addresses this limitation by using an external index mapping keys directly to exact Parquet files and row locations. Readers execute O(1) index lookups and issue parallel ranged reads to fetch only the required bytes without scanning footers or metadata. By running directly against existing data lake files, this architecture eliminates separate storage copies and allows shared analytical data to serve interactive online lookups.


### [Pinner Progression: Better Use-Case Representation Driving Weekly Active User Growth at Pinterest](https://yomu.fyi/post/pinner-progression-better-use-case-representation-driving-weekly-activ.md)
- Company: [Pinterest](https://yomu.fyi/company/pinterest.md)
- Author: Pinterest Engineering
- Published: Jul 27, 2026
- Type: Problem & solution

Pinterest designed the Pinner Progression initiative to shift its home feed recommendation pipeline from short-term engagement optimization toward long-term user retention. Standard ranking systems model user actions as immediate sequential features, failing to capture the evolving lifecycles of distinct user habits. To address this gap, the platform introduced User Interest Clusters (UICs), which cluster a user's engaged Pins in OmniSage embedding space using dynamic cluster counts and stateful lifecycle metadata. Pinterest integrated a UIC-aware penalty directly into Sliding Spectrum Diversity (SSD) scoring by matching Pin embeddings to cluster medoids via cosine similarity. Online experiments demonstrated that balancing use-case representation increased interaction diversity, extended session lengths, and delivered meaningful engagement gains across the platform.


### [NVIDIA Ising Enables Fully Automated Quantum Computer Calibration with Enhanced In-Context Learning](https://yomu.fyi/post/nvidia-ising-enables-fully-automated-quantum-computer-calibration-with.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Tanya Lenz
- Published: Jul 27, 2026
- Type: Announcement

NVIDIA Ising Calibration is an open-source vision language model designed to interpret quantum processor diagnostic outputs and recommend tuning steps. The newly released Ising Calibration 1.5 features a 31-billion-parameter architecture that is 11.4% smaller at BF16 precision than its predecessor. Trained on multi-modality partner data including superconducting qubits, quantum dots, ions, neutral atoms, and electrons on Helium, the model supports both zero-shot reasoning and in-context learning. Evaluated on the QCalEval benchmark, it scores 10% higher on average than comparable open models in zero-shot tasks and improves in-context learning performance by over 86% relative to the prior release. In addition to BF16 weights, the model offers an NVFP4 quantization option suitable for deployment on single consumer GPUs or NVIDIA DGX Spark hardware.


[Newer posts](https://yomu.fyi/posts/page/86.md) · [Older posts](https://yomu.fyi/posts/page/88.md)